EUVIMEDEUVIMED entdecken
Crossref

Performance Evaluation of a Random Forest Classifier for Unauthorized user Detection in Decentralized Network Environments

I.I. Achi, F.O. Ogbunude

International Journal of Modern Science and Research Technology · 2026

DOI: 10.67317/ijmsrt26aug070 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Modern Science and Research Technology, 2026. Verfasst von I.I. Achi und 1 weiteren Autor:innen. Die Arbeit nennt 20 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Unauthorised access constitutes a primary security threat in decentralised networks, where the absence of a single control point multiplies attack surfaces and complicates timely detection. This paper presents a rigorous performance evaluation of a Random Forest classifier engineered to distinguish authorised from unauthorised network activity on the basis of five readily extractable traffic features: packet rate, bandwidth usage, latency, failed login attempts and connection duration. An Isolation Forest component operates in parallel as an unsupervised anomaly detector, supplying coverage for previously unseen (zero-day) behavioural patterns. Experiments conducted on labelled network traffic yield an accuracy of 96.80%, precision of 95.40%, recall of 94.20% and F1-score of 94.80%. By comparison, a conventional signature-based intrusion detection baseline attains only 82.40%, 80.10%, 78.60% and 79.30% on the same metrics. The associated confusion matrix reveals low false-positive and falsenegative counts (18 and 14, respectively, on a balanced test set of approximately one thousand samples). These results establish Random Forest as a practical, interpretable and computationally efficient solution for real-time unauthoriseduser classification in resource-constrained decentralised settings, while Isolation Forest further strengthens resilience against novel attacks.

Zitieren

APA

Achi, I., & Ogbunude, F. (2026). Performance Evaluation of a Random Forest Classifier for Unauthorized user Detection in Decentralized Network Environments. International Journal of Modern Science and Research Technology. https://doi.org/10.67317/ijmsrt26aug070

Vancouver

Achi I, Ogbunude F. Performance Evaluation of a Random Forest Classifier for Unauthorized user Detection in Decentralized Network Environments. International Journal of Modern Science and Research Technology. 2026. doi:10.67317/ijmsrt26aug070

BibTeX
@article{Achi2026,
  author = {Achi, I.I. and Ogbunude, F.O.},
  title = {Performance Evaluation of a Random Forest Classifier for Unauthorized user Detection in Decentralized Network Environments},
  journal = {International Journal of Modern Science and Research Technology},
  year = {2026},
  doi = {10.67317/ijmsrt26aug070}
}
Kostenlos starten →